On the Expert Teacher: A Conversation with

RONALD S. BRANDT On the Expert Teacher: A Conversation with erto th David=l~~~~~~~~~~~~~ _~Berliero "ad time Ihe fild~ \sl'%sz Z~~~~~~~~~~ 4 ErAl :<TI)NAI I.FAI))ERSIIIp

Y our current research is focused on learning how expert teachers think Why? I felt the research on teaching was going in a direction that didn't take seriously a view of the teacher as executive, as high-level decision maker To do that w e had to identify experts: teachers with lots of experi ence who could articulate the internal workings of their minds If we could find out how thev differed from novices, it might tell us something about training needs By novices, do you mean beginners who have some training in education? Novices have either had student teaching or are in their first year of teaching. They have studied pedagogy, but are new to the field. In addition, I wanted to know how the thoughts of experts differed from those we call postulants: educated people who lack training in education, but who want to become teachers. With the shortage of teachers in fields like science and mathematics, there are moves--as in New Jersey-to try to attract such people to enter the teaching order. Yes, I have a view about that. Nobody expects to walk in off the street, no matter how technically able he or she is to fly a Cessna, and fly a 747. No matter how many medical books he may have read, nobody expects to walk in off the street and do brain surgery. But somehow it's fashionable to think you can walk into a classroom of 32 kids-many of whom aren't sure why they are there--and teach them. That's wrong and insulting to teachers. How did you go about finding expert teachers? We began by asking secondary principals to nominate their best math and science teachers. Then we arranged for three independent observers-all of whom were trained teachers and who had talked together about what they thought an expert ought to be like--to observe the teachers on multiple occasions. So your current work is concerned only with secondary mathematics and science teachers? Yes, and for that reason we hadn't the luxurv of a clear outcome measure as we might have if we had used elementary teachers. At the secondary level vou don't have test data, because the standardized tests aren't geared to particular courses, and you can't say as readily what effect a teacher of biologv or chemistry has had on students' science knowledge. You looked into some existing systems for identifying exemplary teachers, didn't you? Yes, and we found that the procedures used often have little to do with the teachers' expertise as a classroom teacher. Take the Teacher of the Year Program. Now, I've talked with numbers of Teachers of the Year-state and national-and there is no doubt in my mind that they're an exceptional group: smart, nice, and articulate. But whether anyv of them could be classed as experts, I don't know. They're not selected on that basis. In one state, a criterion is community activities, including church work In another state they told us a criterion is "really loving children" Now, these are not bad things-you want community involvement and love of kids-but most of us would hesitate to say those are sufficient conditions for expertise. I'm not sure they're even necessary. We found that at the state level thev often rotate the judges every year so as to have maximum involvement in the process. Now, how would vou like to politicize figure skating that way? Or how about having different umpires in baseball ever' year to spread it around? At the national level we found that the final interviews are conducted by representatives of Good Housekeeping I have no idea what Good Housekeeping magazine knows about teaching. You mentioned figure skating. There are other models, then? Absolutely. The model I have become enamored with is cattle judging. There are universities, like Oklahoma and Texas A & M, that actually recruit judges. They get them out of 4H clubs, for example. They hold judging contests, where these college kids will spend four to five hours a dav in judging against established criteria. After the judging they go into the laboratories and slaughter the cattle to find out how close they've come: how much fat on the meat, the bone structure of the cow: they are even trained to judge the beauty of cows. There is a notion of cow beauty that I didn't know existed. But, of course, judging teaching is different. Judging teaching is absolutely no different from judging figure skating, poultry, potatoes, or cows. Each involves making complex decisions with a good deal of subjectivit. "Judging teaching is absolutely no different from judging figure skating, poultry, potatoes, or cows. Each involves making complex decisions with a good deal of subjectivity." I.: OCTOBER 1986

I The difference is practice. It takes 15 years to become an Olympic skating judge. Ten years to be a diving judge. We all marvel at the reliability of the judges as we look at the Olympics every four years. That doesn't just happen. Those people have 20 or 30 years under their belts learning to do thatbut our state departments of education pick new people every year and give them little or no training. Some states pick parents as judges and provide them with half a day of training. What do parents-just because they're parents--know about judging teaching performance? It's an insult to the profession, The Kennel Club requires ten years experience in breeding dogs before you can even submit your name to be a judge. In the Kennel Club you have to post your name in the Gazette and they present it to all members to see if you have the moral fiber to be a judge And why not? There's money and prestige involved-just as there is in teaching "As we learn how expert teachers develop their mental roadmaps, we can use the information to clue in novices. We can say, 'Why not work on your visual scanning?' 'Why not try to automate the roll-taking routine?' Maybe we can help novices concentrate on those things experts do so well." these people I think so highly of have pedagogical knowledge comparable to a chess player's chessboard knowledge?" I'm still in the middle of answering that question, but I am convinced that our early data say yes I'll give you one example. We've discovered that at the start of the school vear, many experienced teachers do something that has been labeled "grooving the kids," which simply means that on opening day you have activities that make sure the students learn very quickly and well that you run the classroom and they don't. Give them assignments, have them fill out things. Now, our experts had never heard of the word "grooving," but they all held the knowledge' that they had to take charge in some way Our novices didn't quite understand that, and the postulants didn't have a clue about how to start off This is a form of domain-specific knowledge chess and on reflection about what the players did. Someone could say, "Remember that Rostropovich game in You mentioned earlier your concern about allowing postulants to 1898?" and everybody would remember the game and all the moves, just as bypass pedagogical training and Before you began your investigation you reviewed the research on when we were kids we might have go directly Into teaching. experts in other fields. said, "Remember the World Series of We went into the study trying to Yes. The modem research starts last year? Remember it was two out in shed some light on policy issues like with republication about 20 years ago the eighth and one man was on?" that. If our work should reveal that of a book about expertise in chess by If you showed these experts a chessboard for just a second and took it there are experts in other fields, we there are experts on pedagogy just as de Groot. In that original study, which took place before World War II, the away, they could recall it completely. might be able to provide the undergirding for the notion of a master researcher found that chess experts Show them an unorganized board and had unique characteristics: their memory for chessboards, for example, was it than you or I. That is, their knowl- We are finding evidence that there they didn't have any better memory of teacher. enormous. The world-class chess players might actually have stored in their way. teachers who are unique in the way edge was organized in a very special are expert teachers; we're finding minds 50,000 images of chessboards Now, bridge players show the same they process information-but we're that they could call upon at any time characteristics. They remember hands also finding that our pool of experts, As they played the game they were of bridge that are organized; they our pool of novices, and our pool of looking for a match between the game don't remember hands that are not. postulants do not separate out easily. they were playing and a model in their Radiologists have special kinds of We have some postulants who seem heads. The minute they had a match, knowledge. So do top-notch problem really sophisticated: who see more they knew how to go about trying to solvers in physics and expert baseball and think more deeply than one or win. That's why some could play 18 players. These people accrue an amazing amount of knowledge in a special- some of our postulants could do well two of our experts. It appears that games simultaneously. Once they discovered which game they were playing, it was relatively easy for them. lar way, and that's the basis for their of summer training under expert suized area and organize it in a particu- very quickly with perhaps five weeks And what was this fabulous memory expertise. pervision. The majority, though, based on? Well, first of all, enormous Having known about that body of would not. They are well meaning, experience-thousands and thousands of chess games. But it was also about the skilled teachers I had simply don't know a lot-which is just psychological research, and thinking bright, and concerned people who based on a language for describing worked with, I asked myself, "Do what you would expect. 6 Eiuc,ziors,i. LnEIisFijp 6 EI)IICATIONAL LEADERKSHIP

Your knowledge of research with experts In other fields enabled you to set up similar experiments? Yes, we think so. For example, we've been trying to see if the memory of our experts is like the memory of chess experts The chess experts are shown a chessboard and even though it's withdrawn quickly, they can recall all the pieces. We show our subjects a slide of a classroom where there's a science or math lesson going on. We flash it on the screen, take it away, and say, "What did you see?" Our novices would say, "I saw a man in a blue gray shirt with a brown tie." "I saw a boy with a book out." "I saw a girl with blond hair." When we asked the experts what they'd seen they invariably answered at a higher level; they described the function Thev'd sav, "The end of the period. They're cleaning up." "It's a hands-on activitr." That's a characteristic of experts---the, have organized knowledge. Experts no longer merely see, they organize what they see into something more meaningful to them. What else have you found about expert teachers? Our experts are bothered by patterns they don't understand. One of our tasks is to show them a slide briefly, then show it again a second and third time Each time we ask them what they have noticed, trying to see how they add information. What they do is focus on abnormality. For example, some of the kids in one slide were looking at a particular area of the classroom.,almost all of our experts noticed it and kept looking for the teacher None of our novices and postulants picked it up. That says to us that our expert teachers have pattern recognition skills similar to expert air controllers, who must know how to read a radar display. If something is out of whack, they spot it. They might not articulate it for a moment, but thev know something is wrong and deal with it. One of the tasks we're experimenting with is designed to see if expert teachers can process several sources of specialized information simultaneously. The literature on experts savs they should be able to. We set up a task in which the teacher has to monitor three T's at once. Over on the left is a group of kids, on the right there's a group of kids, straight ahead is a teacher and another group of kids. The teacher is teaching a lesson on the digestive system using information on an overhead. The question was whether our experts, our novices, and our postulants could process from all three. It turns out our experts were absolutely unbelievable in doing so. Our first expert had no trouble watching all three screens at once and giving statements about things that were going on and, in fact, kept the sound up on all three screens, which I found absolutely impossible. The majority of experts not onlv could do it; thev could make sense from it. The novices that we tested were ven' articulate but didn't have anywhere near the ability to process those three things simultaneously. In one of our tasks we asked our experts, novices, and postulants to look over class materials-graded tests, textbooks, records, and roll book-on a fictitious class they were to take over in two days. The novices and postulants wanted to know about the kids. They read the student information cards carefully because they wanted to know who the bilingual kids were, the learning disabled, the smartest kids, the kids who were going to act out. Our experts almost always skimmed the cards and put them away. They had no interest at all in comments made by other teachers about students because they have learned from many years of experience that kids behave differently in different classrooms. Rather than be biased in any way, the experts avoided the student records. Instead thev focused on the tests: thev wanted to find out what these kids knew. I believe if we had larger samples and could design our studies better, we would find that the novices are concerned with where the holes are. while the experts take off from what kids know. One is a deficit model and the other is a strength model. Now, we have very small samples, so what I am saying is clearly intuitive, but I'm convinced that there is something to it. When they talked about how to start their classes, the novices and postulants wanted to test the students, while the experts preferred to talk with students to find out what they knew. The experts also have such a rich knowledge of subject matter and how it's to be taught that they are more confident. Thev don't need a test. Thev can ask a few questions and say, "Okay, they've an understanding of this aspect of ecology but not that. That's probably where the teacher left off." The experts did another interesting thing. Almost all said they would start off the class by establishing new rules. new procedures, new ways of operating. The novices and postulants wanted to know how the former teacher ran the class so they could continue the system. The experts understood that thev had to take control of the class in their own way It seems that some of the things you've mentioned might be matters of experience rather than real expertse. We don't know vet how to carve out experience from expertise. We're quite sure that a person with many years experience is not necessarily an expert, but all our experts have ten years or so experience. In other words, experience is a necessary but not a sufficient condition for being an expert. OCTOHER 1986?

In psychological studies of expert physics problem solvers the criteria for being an expert are maybe ten years beyond the Ph.D., holding professor status at a university, working on research projects. The novice in such a situation might be a first-year graduate student. There you're talking about experienced-inexperienced and almost everyone is willing to say that's a pretty good way to get expert-nonexpert. None of us would say that in education. One way to find out would be to ask some teachers who are experienced but not regarded as experts to do the same tasks. Absolutely. If we had the money and time we could design it in, but our grant is a relatively small one and my time span is only two and a half years. It's an interesting problem: how does experience change people so that they become experts? Why do some experienced people not become experts? performance as "good," but unless you know what was to have been accomplished and whether it uas accomplished, there is no way to tell whether the teaching was "effective." That's why you mentioned test scores earlier. But when you depend on outcome measures like test scores, you seem to be assuming that the tests we have are adequate measures of what we really want students to learn. I'm only saying that for "effectiveness," there must be an objective standard. Good teaching is based on the norms and values of the observer, but to judge teaching as effective you have to have an outcome. It doesn't have to be a test outcome. You could say that the goal is cooperative learning and an effective teacher is one whose kids cooperate more at the end of the year. there ways for them to use any of what you're learning? One thing that may be helpful is to recognize that novices and experts think very differently. When you ask a figure skater how she switched from inside to outside of the blade, she hasn't got a clue, even though she does it beautifully. She may not even be sure why she does it anymorebecause she mastered that 20 years ago when she was 5 years old In other words, knowledge and action are different things. It's the old story of when you ask a centipede how it walks, it can't walk anymore. That's why supervisors need to ask questions of beginning teachers that give the supervisor clues as to what the teacher is responding to. It's a slow, laborious process Supervisors need to learn how their finest teachers operate and then try to unravel how teachers who are having difficulty think about the same phenomenon I What uses do you foresee for the kind of information you're olth-rin-o think anrt of sunervision is larnino hat for uses do you ery foresee of teacher how to unpack behavior: to open it up the kind of information you're education is the student teaching and try to figure out what the thought gathering? when we place the novice processes were behind it. As research- We have teachers who run the ga- teacher with a "cooperating teacher" ers learn more about the characterismut. Some teach very interesting, un- who is supposedly an expert. We've tics of experts, it may usual tell supervisors lessons. Others seem to excel learned from our studies that some where to look. because of the regularity of their man- expert teachers can't articulate what For example, I'm already convinced agement routines. They get a lot of they do. It turns out that there is a that routines are a key element. From work done and maintain a pleasant growing literature about that. The Re- what we've found so far, one of the environment, but they certainly don't flectivepractitioner, written by Donald major distinguishing characteristics strike us as highly creative people. All Schon, describes experts in many between excellent classroom teachers the experts have lots of routines, by fields who can't tell at all how they do and less expert teachers is that the the way. The advantage of routines for it. Now putting student teachers with experts all have routinized large experts, whether they are chess play- such people-no matter how effective chunks of what goes on. ers or pilots, is that it puts things on they are otherwise-is a big mistake. So supervisors might assist less automatic so their minds are free to do Novices can't learn from them We expert other teachers things. by When helping you know them how to hope our research will provide a bet develop routines? get the papers passed to the right or ter description of how experts act, and Yes. As I think back to teachers I've do the roll quickly or get in from the thereby provide a better language for seen who were less than expert, one playground, you re free to think about describing this hard-to-articulate of the reasons had to do with their lack your lessonsbehavior. of regular procedures. Our experts I want to distinguish, by the way, There are lots of other uses. As we can do a roll check in 30 seconds. between a good teacher and an effect learn how the experts develop their They can get a homework review done tive one Our notion of "effective mental roadmaps, we can use the in- in 75 seconds, while novices might teacher" is someone who meets a formation to clue in novices. We can take 10 minutes Not only that, but as criterion. Our notion of "good teah- say, "Why not work on your visual part of their homework review the er" has no crteria specified. Some scanning?" "Why not try to automate experts can diagnose whether most of might say, for example, the good the roll-taking routine?" And all our the class got it or not, meanwhile also teacher jokes around with the kids for experts have refined their knowledge finding out if certain kids are having a minute or two. When the bell rings by reflective experience. Maybe we trouble at home The expert teacher she or he gets them on task quickly. can help novices learn to concentrate often gets two kinds of information at When a lecture is started, it has an on those things experts do so well once advance organizer. Good questions are asked. There's a review at the end. Many of our members are super.- That leads me to ask whether your Now, there's every reason to judge that visors of one kind or another. Are current research is leading you to I EnrlCATIONAL LEADERSHIP

rethink the findings you reported in past years from the Beginning Teacher Evaluation Study? Two of the major findings of the BTES were that effective teachers had good time management and made sure kids had a high proportion of successful experiences. Another was that they matched the curriculum to the intended outcomes. Our new research supports those earlier findings. An interesting thing, though: our expert teachers all have very high time on task, but once in awhile they surprise us. They run with something for awhile but then stop and don't seem to care about efficiency. In every case they seem to know exactly what they want to accomplish and how to get it done, but there are so many variations in how they operate that it's hard to categorize What does that say about the state of research on teaching? How do you assess what has been accomplished so far and where the field is going? I think we've had an extremely fruitful 25 or so years. I say 25 years because I date the modem research on teaching movement from publication of the first Handbook of Research on Teaching in 1963. That was our equivalent of Gray's anatomy. We have identified a whole host of teacher behaviors and skills that are clearly related to achievement: pacing, structure, monitoring, feedback, certain kinds of questioning behavior. We have research on all sorts of academic climate variables. We know, in other words, what observable teacher characteristics are related to effective teaching. Now we want to go inside teachers' heads and ask them why thev do the things they do For example, as I mentioned, a teacher whose students have more time on task will achieve more. Why then, on a given day, will an expert teacher simply throw the task out the window? All our correlations are based on the notion of stability of teacher characteristics-but every time we do reliability checks we know we're in trouble, because we keep finding the behavior of these expert teachers unstable from day to day and year to year. Why? These are able, experienced people. There's something they are responding to that makes them change a routine that has worked perfectly 30 days in a row. They know something we don't We'd like to find out what.o Refrencas de Groot. A Thought and Choice in hess. The Hague: Mouton Publishers, 1965. Fisher, C W., David C Berliner. N. N. Filbv. R Marliave. L S Cahen, and M. M. Disha- "Teaching Behaviors, Academic Learning Time. and Student Achievement." In C Denham and A Lieberman keds). Time to Lean. Washington. D C., National Institute of Education. 1980 Schon. D. The Reflectse Practioner New York: Basic Books, 1983 David C. Berliner is professor of educational ps-cholog- at the College of Education, Universit- of Arizona. Tucson, AZ 85'21 Ronald S. Brandt is executive editor of ASCD Ell _ A - 3 M S New! LEGAL AND EDUCATIONAL ISSUES AFFECT- ING AUTISTIC CHILDREN by John E. B. Myers, William R. Jenson and William M. McMahon. This practical text offers straightforward guidelines on such topics as the right to an appropriate education, the four steps of 94-142, due process procedures, extended school year programming, behavior management, school records, and filing an adminstrative complaint under 94-142 or Section 504. Sept. '86, about $26. 75 New! ACCESSIBILITY TO EMPLOYMENT TRAINING FOR THE PHYSICALLY HANDICAPPED by Thomas L. Erekson and Anthony F. Rotator. Chapters deal with aspects of job training programs including assessment of physically handicapped students; general accessibility to buildings and equipment; and teaching techniques for students with hearing, visual, orthopedic and consciousness disabilities. Oct. '86, about $22. 75 New! Techniques For Dealing With CHILD SEXUAL ABUSE by Arlene Baxter. This book focuses on the professional's role in the identification, intervention, investigation, treatment, and prevention of sexual abuse. Some of the topics covered are the feelings and attitudes evoked by sexual abuse, physical and behavioral indicators, law enforcement involvement, and community resources. '86, $20.75 New! TEACHING ENGLISH CREATIVELY by John H. Bushman and Kay Parks Bushman. This book provides the busy teacher with an exciting, creative approach to the teaching of English. The opening chapters offer suggestions on effective room decor and atmosphere. Following are chapters on the teaching of writing; literature; thinking skills; the curriculum; and a potpourri of ideas on evaluation, discipline, and communication with administrators and parents. Oct. '86, about $24. 75 New! THE LIFELONG LEARNING EXPERIENCE: An Introduction by John R. erduin, Jr., Harry G. Miller and Charles E. Greer. Three major program thrusts-education of the undereducated, education for career development, and education for leisure-are brought together to emphasize adult lifelong learning. Topics include: adult basic education, English as a second language, and continuing professional education. Sept. '86, about S19.75 New! DEATH EDUCATION AND THE EDUCATOR by James H. Miller and Anthony F. Rotatori. The importance of death education and the need for educators to respond appropriately are examined. This book provides information about the terminally ill student, procedures for dealing with a suicide crisis, and both informal and formal assessment procedures for suicidal behavior. Nov. '86, about $19.75 1u Order direct for fate results Write or call (217) 7 08949- Books saet on approval Postage paid on MasterCard, Visa & prepaid orders * Catalog sent on request nrwlas _

Copyright 1986 by the Association for Supervision and Curriculum Development. All rights reserved.